Professional Documents
Culture Documents
Abstract— One of the goals of the Knowledge Puzzle Project is to automatically generate a domain ontology from plain text
documents and use this ontology as the domain model in computer-based education. This paper describes the generation
procedure followed by TEXCOMON, the Knowledge Puzzle Ontology Learning Tool, to extract concept maps from texts. It also
explains how these concept maps are exported into a domain ontology. Data sources and techniques deployed by TEXCOMON
for ontology learning from texts are briefly described herein. Then the paper focuses on evaluating the generated domain
ontology and advocates the use of a three-dimensional evaluation: structural, semantic and comparative. Based on a set of
metrics, structural evaluations consider ontologies as graphs. Semantic evaluations rely on human expert judgment and finally,
comparative evaluations are based on comparisons between the outputs of state-of-the art tools and those of new tools such as
TEXCOMON, using the very same set of documents in order to highlight the improvements of new techniques. Comparative
evaluations performed in this study use the same corpus to contrast results from TEXCOMON with those of one of the most
advanced tools for ontology generation from text. Results generated by such experiments show that TEXCOMON yields
superior performance, especially regarding conceptual relation learning.
Index Terms— Concept learning, Domain engineering, Knowledge acquisition, Ontology design
—————————— ——————————
retain somewhat of a semantic validity by providing the
means to refer to the original texts.
1 INTRODUCTION However, (semi-)automatic methods for ontology
O ntologies are the backbone of knowledge learning from text should not be considered as a holy
representation for the Semantic Web. In the domain grail that outputs clear and perfect structures. Automatic
of computer-based education, it is believed that knowledge extraction techniques can only provide
ontologies can play a major role in the future of intelligent domain ontology skeletons and more complex building
tutoring systems and eLearning knowledge bases. The steps still require the intervention of human actors.
educational Semantic Web [2] is an initiative to enrich Another issue within the ontology community pertains
learning environments with Semantic Web languages and to the lack of methodologies to evaluate ontologies, be
representations. In this context, ontologies can act as a they built manually or automatically constructed. In fact,
common and reusable knowledge base that training a wide adoption of domain ontologies presupposes a
systems can reuse for learning purposes, provided that means to evaluate the quality, cohesion, domain covering,
such systems adhere to the domain knowledge view and richness of these ontologies.
expressed in the ontology. This paper presents TEXCOMON, a knowledge
In fact, knowledge is never a fixed entity: it evolves extraction tool produced within the Knowledge Puzzle
with new discoveries and usages. In order to keep Project. TEXCOMON provides a solution for the
ontologies updated with such advances, automatic aforementioned issues by generating semi-automatic
methods to build them and extend or update them must domain ontologies from texts, and by offering a clear
be set up. The dynamic nature of knowledge implies that evaluation methodology to analyze ontologies from three
manual methods used to build domain ontologies are not perspectives: the structural, semantic and comparative
scalable: they are time and effort consuming and dimensions. The goal of this paper is to present
represent knowledge as a set structure established at the TEXCOMON and the evaluation methodology used to
time the ontology was conceived and built. assess the generated domain ontologies. It is organized as
In order to minimize these drawbacks and avoid the follows:
tremendous effort of consistently starting over again, Section 2 briefly presents state-of-the-art elements in
automatic methods for domain ontology building must be the domain of automatic ontology building and
adopted. Various domain documents can be used as a evaluation with an emphasis on the evaluation
source of knowledge. Using domain texts to capture the methodologies. Section 3 details the software suite
view of a certain community can help preserve a TEXCOMON and explains ontology learning from
consensus among community members. Since the domain texts. The remainder of the paper focuses on a
ontology “emerges” from texts, it is possible to explain methodology to evaluate the generated ontology based
the presence of a particular concept, property, instance or on structural measures (Section 4) and on comparative
attribute. Hence, ontology learning from texts can help measures with one of the most advanced system in
————————————————
domain ontology generation: TEXT-TO-ONTO (Section
4.4). A semantic evaluation is also performed. Finally,
The authors are with the University of Quebec at Montreal, Pavillon
Sherbrooke 200, rue Sherbrooke ouest, local SH-5720, Montreal,
QC H2X 3P2, Canada. E-mail: {zouaq.amal, nkambou.roger}@uqam.ca.
xxxx-xxxx/0x/$xx.00 © 200x IEEE
2 IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, TKDE-2007-09-0444
obj - object
dobj - direct object In this example, the term T has one input link (rcmod:
iobj - indirect object relative clause modifier) and two output links, namely
pobj - object of preposition nominal subject and direct object. This pattern includes an
additional constraint: the nsubj link must point towards a
agent – agent relative clause modifier (that, which, etc.). The
abbrev - abbreviation modifier transformation process creates a new relationship labeled
sdep - semantic dependent as T between the source of the rcmod link (X) and the
appos – appositive destination of the dobj link (Z). An occurrence of this
pattern can be found in the following sentence: “The
Other terminological patterns rely on output links prescription specifies the activities that use the content objects
from a source term T and need a small structural in the package”. In this sentence, the typed dependency
transformation, by composing a new term from T and its network is:
destination. These links are the following: amod - adjectival nsubj(specifies-3, prescription-2)
modifier and nn - noun compound modifier, as shown below. dobj(specifies-3, activities-5)
nsubj(use-7, that-6)
T U rcmod(activities-5, use-7)
dobj(use-7, objects-10)
amod or nn nn(objects-10, content-9)
Moreover, the aggregation operation involving an prep_in(objects-10, package-13)
“amod” link yields the creation of a taxonomical link
between the source term T and the newly composed term The pattern analyzer creates a semantic relationship
U T. For instance: “activities – use – content objects” from the above pattern.
It also finds an instance of the subject-verb-object pattern
System Intelligent described above and creates the relationship
“prescription – specifies – activities”.
amod The whole process results in a set of semantic
The transformation results in a new composite term, relationships constituted by a source term, a destination
i.e., “Intelligent System” and a taxonomic link: is-a term and a label. Each of these semantic relationships is
(Intelligent System, System). stored in a property “Relation”, which is linked to its
source concept. Semantic relationships contain a pointer
3.4. Identifying Relational Patterns towards their originating sentences. This allows for
Relational and lexico-syntactic patterns, coupled with effective indexing by sentence, paragraph and entire
Java methods, transform a grammatical structure into a document.
“semantic” one. As with previous terminological
patterns, they also rely on detecting sub-graphs in typed
3.5. Example of a Semantic Analysis
dependency networks. An example of a well-known Figure 3 shows a complete example of a
pattern is the subject-verb-object pattern, schematized transformation process based on lexico-syntactic patterns.
below: The parsed sentence (“An asset is a content object that
will not use the SCORM API, but that can still be used for
an activity”) is decomposed into a set of typed
nsubj dobj dependencies.
X T Y
The transformation results into the following triple: Step 1: Removing determiners
nsubj(object-6, asset-2)
cop(object-6, is-3)
ZOUAQ AND NKAMBOU: EVALUATING THE GENERATION OF DOMAIN ONTOLOGIES IN THE KNOWLEDGE PUZZLE PROJECT 5
The following section describes the conversion of implements an algorithm to identify correct acronyms,
concept maps into a domain ontology. which are stored as terms associated with the current
concept.
3.7. Converting Concept Maps into a Domain Most extracted classes belong to primitive classes.
Ontology However, some defined classes can also be detected
Domain concept maps act as skeletons on which through the “abbrev” link. Each concept and its acronym
domain ontologies are built. This process implies are defined as equivalent classes, as shown below.
determining classes, relationships, attributes and <owl:Class rdf:ID="runtime_environment">
instances in the concept maps. <rdfs:subClassOf rdf:resource="#environment" />
<owl:equivalentClass>
3.7.1. Defining Classes
<owl:Class rdf:ID="RTE" />
Extracting ontological classes (concepts) from concept </owl:equivalentClass>
maps is performed by detecting the presence of high </owl:Class>
density components, which indicate the importance of a At the current time, the Knowledge Puzzle lacks the
given concept. In the Knowledge Puzzle, a term is ability of handling anaphors and cannot process
considered a concept when it is linked to other domain antecedents such as ‘’reference model’’ and ‘’the model’’
terms through a number of semantic relationships. This in the following text: “SCORM is a reference model […].
number can be parameterized according to the corpus The model …”
size and the human experts’ goals.
A single concept in a text can be expressed in different 3.7.2. Defining Relationships
ways. The Knowledge Puzzle can recognize the base form Basically, all verbal relationships between pairs of
of a concept through stemming. TEXCOMON uses a Java classes are considered as potential ontological
version of the Porter Stemmer [34] to produce the stem relationships. The relationships generated include simple
associated with each concept. For example, the words object properties such as:
“stemmer", "stemming" and "stemmed" have the same <owl:ObjectProperty rdf:ID="may_need">
root: “stem-”. This is particularly useful as it allows <rdfs:domain
recognizing the plural forms of nouns and certain rdf:resource="#training_resources" />
conjugated verbs. Another way of expressing concepts is <rdfs:range rdf:resource="#metadata" />
through acronyms (e.g., ‘’SCO’’ stands for ‘’Sharable </owl:ObjectProperty>
Content Object’’). Although the Stanford University An object property can also take the shape of a blend
Parser outputs acronym links as typed dependencies, this of classes in its range or domain, as shown below:
feature is not always reliable. Hence, the TEXCOMON <owl:ObjectProperty rdf:ID="describes">
ZOUAQ AND NKAMBOU: EVALUATING THE GENERATION OF DOMAIN ONTOLOGIES IN THE KNOWLEDGE PUZZLE PROJECT 7
SCORM and 1.
LMS The functions (ONTO-EVALUATOR library) were
Runtime Environment implemented to perform different metrics computations
Metadata based on the exact formulas described in [1]. Jung’s
SCORM Content Packaging Betweenness algorithm [14] was also used to directly
Activity calculate the betweens measure.
Content Organization 4.3.1 Class Match Measure (CMM)
API
The Class Match Measure (CMM) evaluates the
PIF
coverage of an ontology for the given sought terms.
As mentioned above, with the TEXCOMON approach, Given the input sought terms, the ONTO-
experts must assign a value to a parameter that represents EVALUATOR searches the classes in the ontology, to
the out-degree of a concept. During the experiment, four determine if the sought terms correspond exactly to
domain ontologies were generated from the same corpus. ontological classes (exact match) or if they are included in
These ontologies correspond to different values of a given the label of one or many classes (partial match).
parameter I. In other words, I=2, which outputs the Figure 5 shows the evolution of the CMM values for
ontology KP-2; I=4, which outputs the ontology KP-4; etc. the different corpuses by taking into account partial and
The second step of the experiment consists of exact matches.
performing the same kinds of measures on the ontology
generated by TEXT-TO-ONTO and comparing the results 1.5
in terms of concept existence, richness and
interconnection levels. 1
Seven sub-corpuses are derived from the 36 documents
used to assess the evolution of the different metrics when
0.5
new domain documents are added to the previous corpus
in an incremental manner (Table 3). For example, Corpus
2 contains Corpus 1, to which four new files were added. 0
terms. 1
This first vision was slightly modified by considering 0.8
an initial set of key search terms as being representative 0.6
of a domain, in an attempt to identify if the generated
0.4
ontology includes these terms as classes and to assess
how many of these terms are interconnected and richly 0.2
described by attributes. 0
The structural metrics are the Class Match Measure Corpus1 Corpus2 Corpus3 Corpus4 Corpus5 Corpus6 Corpus7
(CMM), the Density Measure (DEM), the Betweenness
KP-2-CMM KP-4 - CMM KP-6 - CMM KP-8 - CMM
Measure (BEM) and finally the Semantic Similarity
Measure (SSM). A total score is then computed from all Fig. 6 : CMM resulting from exact match
these measures. This score can be used to rank the
ontology with respect to the given search terms. The With Corpus no 7, KP-2, KP-4 and KP-6 achieve
values of all metrics and final scores are set between 0 identical results. However, KP-8 offers a poorer
ZOUAQ AND NKAMBOU: EVALUATING THE GENERATION OF DOMAIN ONTOLOGIES IN THE KNOWLEDGE PUZZLE PROJECT 9
4.3.2 Density Measure (DEM) KP-2-BEM KP-4 - BEM KP-6 - BEM KP-8 - BEM
The density measure expresses the degree of detail or
the richness of the attributes of a given concept. The Fig. 8 : BEM Evolution across corpuses and thresholds
underlying assumption is that an adequate representation
of a concept must provide sufficient details about this
4.3.4 Semantic Similarity Measure (SSM)
concept. The density measure includes the number of
subclasses, of inner attributes, of siblings and of The last measure, the Semantic Similarity Measure
relationships with other concepts. (SSM), computes the proximity of the classes that match
Figure 7 shows DEM evolution across corpuses and the sought terms in the ontology. As stated by Alani and
thresholds. Brewster [1], if the sought terms are representative of the
domain, then the corresponding domain ontology should
1.5 link them through relationships (taxonomic or object
properties). Failing to do so may indicate a lack of
1 cohesion to represent this domain knowledge.
The SSM is based on the shortest path that connects a
0.5 pair of concepts. As shown in Figure 9, correlations exist
between the text volume and the SSM. The SSM never
0 decreases, regardless of the threshold value. In general, a
high threshold value yields a poorer performance of the
SSM value. However, with larger corpuses, high
thresholds become more appropriate.
KP-2-DEM KP-4 - DEM KP-6 - DEM KP-8 - DEM
1.5
Fig. 7: DEM Evolution across corpuses and thresholds
1
The DEM tends to increase proportionally to the
number of concepts. These variations result from the 0.5
richness of information in the new corpus. For example,
Corpuses 6 and 7 probably add many new relationships, 0
explaining the significant increase, especially when
threshold = 2.
KP-2-SSM KP-4 - SSM KP-6- SSM KP-8 - SSM
4.3.3 Betweenness Measure (BEM)
The BEM calculates the betweenness value of each Fig. 9: SSM Evolution across corpuses and thresholds
search term in the generated ontologies. Betweenness
indicates the extent to which a concept lies on the paths
between others. The underlying assumption is that the As previously stated, considering only identical
centrality of a class in an ontology is important. A high matches has a significant impact on this metric (see
betweenness value shows the centrality of this class. As in Figure 10).
ActiveRank [1], Onto-EVALUATOR uses the BEM 1.2
provided by JUNG [14]. This algorithm calculates the 1
number of shortest paths that pass through each concept 0.8
in the ontology, considered as a graph. A higher 0.6
betweenness value is assigned to concepts that occur on 0.4
many ‘’shortest paths’’ between other concepts. 0.2
A reasonable number of relationships must be retained 0
in order to have an significant BEM. Figure 8 suggests
that again, thresholds 2 and 4 seem to be the best options
available.
KP-2-SSM KP-4 - SSM KP-6- SSM KP-8 - SSM
In that case, the exact match leads to very similar generated by the TEXT-TO-ONTO ontologies are
results for KP-2, KP-4 and KP-6, especially with the most obtained with the same metrics as presented above. Not
extensive corpus (no 7) where identical results are all metrics results are shown here due to a lack of space.
obtained. This is not the case if partial and exact matches The generation process followed by TEXCOMON is
of input terms are adopted (Note the differences between identical to the one explained in Section 3.
Figures 9 and 10).
Finally, an overall score can be computed, based on 4.4.1 Overall Score
these four metrics. The weights assigned to each of the The total score of an ontology can be calculated once
metrics can be either the same or different. This overall the four measures are applied to the entire set of
score will be explained in the following section. generated ontologies. The global score is calculated by
summing up all the measured values and taking into
4.4 Comparative Evaluation account the weight of each measure. Varying weights can
The comparative analysis involves the generation of an be helpful to determine the relative importance of each
ontology with TEXT-TO-ONTO using the same corpuses. measure for ranking purposes.
Unlike TEXCOMON, TEXT-TO-ONTO has no parameters By assigning identical weights to all metrics (Table 4,
related to the out-degree of concepts. However, other Figure 11), TEXCOMON clearly outperforms TEXT-TO-
parameters can be taken into account, more particularly ONTO in all corpuses. KP-8 is the only ontology that
the support given to the association rules generated by yields a lower score as compared to those of TEXT-TO-
the algorithm. ONTO.
To clarify this notion of support, it is important to first Table 4. Ontology Overall Scores and Ranks on the
define association rule learning. This type of learning can Largest Corpus (no 7) with identical weight (0.25) for all
detect the items (in this case: terms), which co-occur metrics.
frequently and extract rules that connect such items. The Ontology Score Rank
support of an association rule is the percentage of groups KP2 1 1
(in this case: documents) that contain all the items of the
KP4 0.41 2
rule.
Two ontologies were generated with TEXT-TO-ONTO KP6 0.34 4
(TTO-1, TTO-2) using a total of seven corpuses. Two KP8 0.26 6
supports were considered: one of 0, indicating that any TTO-1 0.38 3
processed rule is considered valid (i.e., TTO-1) and the TTO-2 0.29 5
other of 0.1 (i.e., TTO-2). In other words, with TEXT-TO-
ONTO, each corpus enables the generation of two
1.2
different ontologies.
This experiment shows that even a 0.1 support value 1
Note that TEXCOMON discovers more super-classes domain ontologies reflect the domain knowledge. It is
and siblings for these two concepts (especially for KP-2 believed that such an evaluation can only be performed
and KP-4). Also, note the disproportion between the by domain experts. Two experts were asked to analyze
number of properties in TTO-1 and TTO-2 and between the resulting ontologies and to remove all inappropriate
TTO-1, TTO2 and KP ontologies. Again, the 172 relations concepts, attributes and relationships. The percentage of
generated in TTO-1 turn into a null value in TTO-2. pertinent concepts and relationships was then calculated
Table 10 shows two excerpts of the generated following the assessment by each expert. This operation
relationships regarding the terms SCO and asset in was performed on the TEXCOMON and the TEXT-TO-
TEXCOMON. ONTO ontologies.
Table 10. An excerpt of the relationships generated for The following table summarizes the evaluation of the
the terms asset and SCO in TEXCOMON ontologies. TEXCOMON ontologies, using the mean scores for
Terms Relationships Generated relevancy, as expressed by the two experts. As shown in
can_be_used_for – activity the table, results are promising. The same procedure is
will_not_be_included_as - physical_files then repeated with TEXT-TO-ONTO ontologies (Table
Asset 12).
is_taken_away_from - learner
is_basic_building_block_of - training_resources Table 12. Mean scores for relevant data generated by
does_not_communicate_to - LMS
both solutions in %).
is_collection_of - Asset
Primitive Defined Hierarchical Conceptual
are - responsibilities Classes Classes Relationships Relationships
is_tracked_by - LMS Ontology
must_behave_within - runtime_environment KP-2 86.65 55.55 84.3 80.08
may_initiate - communication KP-4 90.84 100 84.83 89.65
must_able_to_consistently_find - API_Instance KP-6 90 100 77.1 91.15
SCO
can_be_described_with - Metadata KP-8 90.32 100 75.28 93.12
to_find - API_Instance TTO-1 73.06 n/a 47.53 0.31
is_intended_to_be_delivered_in - LMS TTO-2 73.06 n/a 47.53 53.03
may_communicate_to - LMS
According to the results of the semantic evaluation,
terminates - communication
TEXCOMON ontologies yield a superior performance.
is_required_to_adhere_to - requirements
This is even more significant in conceptual relationships
finds - API_Instance
learning, one of the strengths of TEXCOMON, as well as
An excerpt of the relationships generated by TEXT-TO- one of the most challenging tasks in text mining. The
ONTO was also investigated regarding the term asset results of the semantic evaluation confirm those of
(Table 11). The same kind of results was found for the previous structural and comparative analyses. However,
term “SCO” and was not shown below. it cannot be disregarded that this novel algorithm takes
Table 11. An excerpt of the incident edges for the
concept ‘’asset’’ in TTO- 1. much time to process the entire corpus (a total of
approximately 5 hours) while TEXT-TO-ONTO outputs
Term Relationships Generated results much more quickly.
defaultProperty1,068 - launch
defaultProperty1,971 - train
defaultProperty3,494 - lm 5 RESULT ANALYSES AND DISCUSSION
defaultProperty1,912 - refer From the results of this experiment, it can be seen that
Asset defaultProperty690 - creat TEXCOMON ontologies yield superior performance than
defaultProperty2,525 - metadata those of TEXT-TO-ONTO, especially when compared to
defaultProperty1,631 - learner TTO-2.
defaultProperty3,066 - docum Revealing results are found when varying weights,
defaultProperty472 - experi exact matches and by observing their impact on the
defaultProperty1,346 - packag overall ontology scores. The following critical questions
The problems with TEXT-TO-ONTO are the following: must be considered when modifying such parameters:
1) it does not extract relationship labels between concepts First, and most importantly, is it possible to obtain a more
in association rule learning and 2) it fails to keep the compact ontology that preserves the performance or the score
complete label of a concept, storing only its stem (in the levels of KP-2?
OWL ontology). TEXCOMON extracts both types of Second, which are the most important metrics
labels (complete labels and stems) and it also extract according to the domain, the goal and the needs?
labels for relationships. If the answer to the first question is affirmative, then a
more compact ontology should be favored over one that
4.5 Semantic Evaluation is less compact, since it includes more richly
The third component of this analysis, the semantic interconnected concepts while preserving the sought
evaluation, relies on human experts to assess the validity domain terms. For example, in Table 6, KP-4 should be
of the ontology. The semantic evaluation is aimed at chosen whereas Table 7 indicates that KP-6 is the best
detecting to what degree, and how well, the generated ontology: its score is identical to that of KP-2 and KP-4,
ZOUAQ AND NKAMBOU: EVALUATING THE GENERATION OF DOMAIN ONTOLOGIES IN THE KNOWLEDGE PUZZLE PROJECT 13
but it is much more compact than the latter two. KP-4, KP-6 and KP-8.
Table 8 depicts comparative data for the output of both In summary, there is no single right way of evaluating
ontologies in terms of number of concepts and ontologies. However, certain lessons can be retained from
relationships (taxonomic and non-taxonomic). Note that: this experiment:
TEXCOMON results can be parameterized. In fact, any In the absence of a gold standard evaluation method
given ontology designer may be interested in larger or for a given domain ontology, building a new one is not
more condensed ontologies and should be given the always possible. In such a case, another ontology
opportunity of fine-tuning results. evaluation method must be undertaken;
With TEXCOMON, a threshold increase is Comparing the generated domain ontology with
proportional to a decreasing number of concepts and others generated by state-of-the-art tools can be beneficial
relationships. to highlight the added value of the new tool or platform.
A revealing aspect appears in the number of non- This confirms the interest of the comparative evaluation
taxonomic links in TTO-2 (n=33) compared to TTO-1 as proposed in this paper;
(n=5683). This tremendous decrease pertains to the Evaluating an ontology from a structural point of view
support of 0.1 used by TTO-2, meaning that TTO-1 can also be interesting as shown in [1]. Comparing this
created relationships corresponding to association rules structural evaluation as we did with other generated
whose support is inferior to 0.1. Although such ontologies, is meaningful. However, this comparison
relationships contributed to the illusion of an improved should be accompanied with stringent semantic
performance of TTO-1, especially with SSM measures, evaluations.
they are actually meaningless and offer no added value
for ontological relationships.
Given that the majority of relations come from
6 CONCLUSION
detecting association rules, TEXT-TO-ONTO rarely This paper presented domain ontologies generated in
extracts labels for relations between concepts, unlike the Knowledge Puzzle Project through TEXCOMON, by
TEXCOMON. This lack of labels translates into a label as focusing on ontology evaluations. Built on Alani and
"defaultProperty" and, subsequently, it is rather difficult Brewster metrics [1], a complete structural evaluation was
to assign meaning to this type of relationship. presented. A comparative evaluation was also proposed
In the end, what kind of results can be deduced from with state-of-the-art ontology generation tools and more
the above statements? In the case of the structural specifically with TEXT-TO-ONTO [20]. The goal was to
evaluation, TEXCOMON offers the possibility of compare the ontological outputs in terms of concepts,
calibrating thresholds according to ontology designers’ attributes, hierarchical and non-taxonomic relationships.
needs and goals. Given a set of search terms which are Overall scores were computed, based on structural
considered important domain concepts: metrics, in order to provide a glimpse into how ontologies
Threshold calibration can be performed by taking into can be compared, using various parameters. Finally, a
account CMM, if the most important feature is the partial semantic evaluation was performed.
or complete match of search terms as ontological When compared with TEXT-TO-ONTO, TEXCOMON
concepts. produced more interesting results in terms of concepts
If the important feature consists of generating richly and relationships. However, this does not mean that there
described concepts with a significant number of attributes is no room for improvement. In fact, a lot of noise is
and relationships, then the density measure should have generated by lexico-syntactic patterns and their
a larger weight in the overall evaluation. associated methods. Further work must be performed in
If the important feature consists of finding richly order to enhance these patterns. Efforts must also be
interconnected concepts in order to make them central to invested in order to reduce the overall processing time of
the ontology, semantic similarity and betweenness should documents. It was noticed that Protégé database projects
be considered. tend to slow down as the number of instances increases.
Our opinion is that all the measures are important. In Moreover, it seems that the OWL Java API used in the
general, and if we take into account the overall score, Knowledge Puzzle Project could also be improved in
ontologies KP-2 and KP-4 seem satisfactory, given the terms of processing time. Finally, other experiments must
corpus size. However, one should bear in mind that be performed to determine how much of the knowledge
structural evaluations alone do not suffice and that they contained in the documents is actually retrieved by
can be misleading when used alone. For instance, the TEXCOMON. For now, solutions to these specific
structural evaluation awarded a better score to TTO-1 in challenges have yet to be found.
comparison with TTO-2 while TTO-2 generated much There is no single best or preferred approach to
more significant results in terms of semantics. The ontology evaluations: the choice of a suitable approach
semantic evaluation confirmed the results of the must reflect the purpose of the evaluation, the application
structural assessment. The results of this analysis point in which the ontology will be used and certain specific
towards the choice of Ontology KP-4, which offers the aspects of the ontology under evaluation. In our opinion,
best relevancy rate for pertinent classes and hierarchical future work in this area should focus particularly on
relationships. As far as conceptual relationships are automated ontology evaluations, a necessary prerequisite
concerned, there are no significant differences between for the fruitful development of automated ontology
14 IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, TKDE-2007-09-0444
processing techniques in order to solve a number of [19] A. Maedche and S. Staab, “Measuring similarity between
problems, such as ontology learning, population, ontologies,” Proc. of CIKM 2002, LNAI vol. 2473, pp. 251 – 263,
mediation and matching. Springer-Verlag, London, UK, 2002.
[20] A. Maedche and S. Staab, “Ontology Learning for the Semantic
REFERENCES Web,” IEEE Intelligent Systems, 16 (2): 72-79, 2001.
[21] J. D. Novak, and A. J. Cañas, “The Theory Underlying Concept
[1] H. Alani and C. Brewster, “Metrics for Ranking Ontologies,”
Maps and How to Construct Them”, Technical Report IHMC
Proc. of 4th Int. EON Workshop, 15th Int. World Wide Web Conf.,
CmapTools 2006-01, Florida Institute for Human and Machine
Edinburgh, 2006.
Cognition, 2006.
[2] L. Aroyo, and D. Dicheva, “The New Challenges for E-learning:
[22] M. Poesio and A. Almuhareb, “Identifying Concept Attributes
The Educational Semantic Web,” Educational Technology &
Using A Classifier,” Proc. of the ACL Workshop on Deep Lexical
Society, 7 (4), 59-69, 2004.
Acquisition, pp. 18-27, Ann Arbor: Association for
[3] J. Brank, M. Grobelnik, and D. Mladenić, “A survey of ontology
Computational Linguistics, 2005.
evaluation techniques,” Proc. of the 8th Int. multi-conference
[23] R. Porzel, and R. Malaka, “A task-based approach for ontology
Information Society IS-2005, Ljubljana, Slovenia, 2005.
evaluation,” Proc. of ECAI 2004 Workshop Ont. Learning and
[4] C. Brewster, H. Alani, D. Dasmahapatra, and Y. Wilks, “Data
Population, pp. 9–16, 2004.
driven ontology evaluation,” Proc. of International Conference on
[24] R. Rada, H. Mili, E. Bicknell, and M. Blettner, “Development and
Language Resources and Evaluation (LREC), pp. 26–28, 2004.
application of a metric on semantic nets,” IEEE Transactions on
[5] P. Buitelaar, D. Olejnik, and M. Sintek, “A Protégé Plug-In for
Systems Management and Cybernetics, 19(1):17–30, 1989.
Ontology Extraction from Text Based on Linguistic Analysis,”
[25] P. Resnik, “Semantic similarity in a taxonomy: An information-
Proc. of the 1st European Semantic Web Symposium (ESWS),
based measure and its application to problems of ambiguity in
Heraklion, Greece, 2004.
natural language,” Journal of Artificial Intelligence Research,
[6] S. Cederberg and D. Widdows, “Using LSA and Noun
11:95–130, 1999.
Coordination Information to Improve the Precision and Recall
[26] A. Schutz, and P. Buitelaar, “RelExt: A Tool for Relation
of Automatic Hyponymy Extraction,” Proc. of the Conference on
Extraction in Ontology Extension,” Proc. of the 4th International
Natural Language Learning (CoNNL), Edmonton, Canada, 2003.
Semantic Web Conference, pp. 593-606, 2005.
[7] P. Cimiano, “Ontology Learning and Population from Text,”
[27] SCORM (Sharable Content Object Reference Model). Retrieved
PhD thesis at the Universität Karlsruhe (TH), 2006.
September 5th, 2007 from
[8] P. Cimiano, J. Volker, and R. Studer, “Ontologies on Demand? –
http://www.adlnet.gov/scorm/index.aspx
A Description of the State-of-the-Art, Applications, Challenges
[28] M. Shamsfard, and A. Barforoush, ”The state of the art in
and Trends for Ontology Learning from Text,” Information,
ontology learning: a framework for comparison,” Knowl. Eng.
Wissenschaft und Praxis 57 (6-7): 315-320. October 2006.
Rev. 18 (4): 293-316, 2003.
[9] P. Cimiano, A. Hotho and S. Staab, “Comparing Conceptual,
[29] http://sourceforge.net/projects/texttoonto
Partitional and Agglomerative Clustering for Learning
[30] UIMA (2007). Retrieved August 28th, 2007 from: http://uima-
Taxonomies from Text,” Proc. of the European Conference on
framework.sourceforge.net/
Artificial Intelligence (ECAI’04), pp. 435-439, IOS Press, 2004.
[31] J. Völker, P. Hitzler, and P. Cimiano, “Acquisition of OWL DL
[10] M-C. De Marneffe, B. MacCartney and C.D. Manning,
Axioms from Lexical Resources,” Proc. of the 4th European
“Generating Typed Dependency Parses from Phrase Structure
Semantic Web Conference (ESWC'07), pp. 670-685, Springer, 2007.
Parses,” Proc. of the 5th Conference on Language Resources and
Evaluation, pp. 449-454, Genoa, Italy, 2006. Amal Zouaq received her Ph.D. (2008) in
[11] E. Frank, G.W. Paynter, I.H. Witten, C. Gutwin, and C.G. Nevill- Computer Science from the University of
Manning, “Domain-specific key phrase extraction,” Proc. of the Montreal. She is now a researcher at the GDAC
16th International Joint Conference on Artificial Intelligence, pp. laboratory (Knowledge Management Lab,
University of Quebec at Montreal (UQAM)). She
668-673, San Francisco, USA, 1999. is also a lecturer in the Computer Science
[12] P. Hayes, T. Eskridge, R. Saavedra, T. Reichherzer, M. Mehrotra, Department at the University of Quebec at
and D. Bobrovnikoff, “Collaborative Knowledge Capture in Montreal. Her research interests include
knowledge representation and extraction,
Ontologies,” Proc. of the 3rd International Conference on Knowledge
Semantic Web, data mining, ontology engineering and computational
Capture (K-CAP’2005), pp. 99-106. ACM press, 2005. linguistics.
[13] M. Hearst, “Automatic Acquisition of Hyponyms from Large
Text Corpora,” Proc. of the Fourteenth International Conference on Roger Nkambou is currently a Professor in
Computer Science at the University of Quebec
Computational Linguistics, pp. 539–545, Nantes, 1992. at Montreal, and Director of the GDAC
[14] http://jung.sourceforge.net/ (Knowledge Management Research)
[15] http://kaon.semanticweb.org/ Laboratory (http://gdac.dinfo.uqam.ca). He
received a Ph.D. (1996) in Computer Science
[16] D. Klein and C.D. Manning, “Accurate unlexicalized parsing,”
from the University of Montreal. His research
Proc. of the 41st Meeting of the Association for Computational interests include knowledge representation,
Linguistics, pp. 423 – 430, Sapporo, Japan, 2003. intelligent tutoring systems, intelligent
[17] D. Lin and P. Pantel, “Discovery of inference rules for question software agents, ontology engineering,
student modeling and affective computing. He
answering,” Natural Language Engineering, 7(4):343–360, 2001. also serves as member of the program committee of the most
[18] A. Lozano-Tello and A. Gomez-Perez, “Ontometric: A method important international conferences in Artificial Intelligence and
to choose the appropriate ontology,” J. Datab. Mgmt., 15(2):1–18, Education.
2004.